In the patent EP 1 047 518, it is shown that a high speed compaction (HVC) process together with an agglomerated spherical metal powder offer distinct advantages.
Bos et al in Powder Metallurgy vol 49, no 2, pp 107-109 discloses a process where the powder first is compacted traditionally and pre-sintered to burn off the lubricant. The parts are then compacted again using HVC and finally sintered traditionally. It is also stated that multilevel HVC has the potential to attract a market segment not previously feasible for PM.
WO 03/008131 discloses a process wherein in one embodiment a multilevel preform is inserted into a cavity of a tool and compacted by HVC. In another embodiment particulate material is inserted into a cavity and compacted to a preform. The pre-form is then compacted by HVC.
US 2008/0202651 discloses a method comprising the steps pre-compacting metal powder, pre-sintering the metal powder at 1000-1300° C., and compacting the pre-form by HVC.
There is plenty of room for an improvement regarding manufacture of multilevel components with HVC. This is due to the fact that the high speed of the ram makes it difficult or even impossible for the powder material to flow around in the cavity and thereby fill up all volume in a tooling die with a complicated shape such as a multilevel part. The filling of the cavity in the tool is in traditional compactions made so that a shoe is brought over the cavity, filling up the tool up to the upper level of the tool. In a conventional tooling set there are also often internal parts, see FIG. 1, which are moving up or down during the pressing operation, thereby creating the multilevel pressed part. This is in practice not possible to do during HVC or similar methods.
Another room for improvement concerns the upper limit of densification. Due to the adiabatic effect, described in the patent EP 1 047 518, it is possible to reach very high densities with HVC, way over the conventional pressing technique. However, due to the need for debinding a binder such as a hydrocolloid it is necessary to stop the densification at a certain upper limit to allow the binder to evaporate during this step.
Other undesired phenomena can also occur in the state of the art at extremely high densities with the binder incorporated such as blisters in the surface.
A further area where there is a room for improvement is the tolerances of a pressed multilevel part, which at the same time has full density and the associated desired mechanical properties.
A further problem in the state of the art is that the density of a uniaxially compressed part differs in the part, due to factors such as friction against the wall of the tool.
It is well known in the art that it so far has not been possible to use high speed compaction to compact powder materials with a grain size of less than 1 mm to multilevel parts.